ECG Spectrogram and Modified InceptionNet for Cardiovascular Disease Classification

An increasing number of people are developing cardiovascular illnesses. Concern over how serious these illnesses are is growing. The electrocardiogram (ECG) is a commonly used diagnostic tool for many medical conditions. Regretfully, obtaining expert techniques to analyze large amounts of ECG data consumes much too much medical time and resources, and existing techniques continue to perform poorly across multiple recording sessions. However, the advancement of deep learning approaches for ECG signal categorization has reached cardiologist performance level in recent years, whereas manual examination of ECG data is laborious and challenging for novice physicians. A method to categorize the cardiovascular diseases is introduced in this study. A modified version of the Inception model is used to classify the spectrograms of ECG signals. Experiments indicate that the suggested approach achieves acceptable performance that is higher than previous efforts.